In September, I gave a talk entitled ‘AI is going to kill us all, but meanwhile can it get us a better Kingston?’ Created entirely with AI-generated content, it aimed to look at both the global and local implications, the risks and benefits, and whether to surrender to the already present, near-ubiquitous usage given a soon-enough and probably inevitable sentient AI.

My talk, though, had a very early genesis. As an open-mouthed, wide-eyed 11-year-old watching the film version of HG Wells’s The Time Machine in 1960, I fell first in love with sci-fi and then with Yvette Mimieux, who played Weena. It might have been the other way around. I have lived in a parallel universe of sci-fi-infused existence ever since. Ever the pseudo-philosphical sci-fi flâneur, my intellectual bar is very low, and I will watch or read anything that sets up dystopias or, indeed, utopias. Although I will confess that the so-called MCU (Marvel Comics Universe) has all but extinguished my affection for superheroes.

Within that oeuvre, I have been following the philosophical trajectory of robot ethics since I read Isaac Asimov’s books in the 1960s featuring robots with ‘positronic’ brains and their Three Laws to ensure that the robot does not turn against its creators. Meanwhile in another branch of life-long curiosity, I have since attending art school (so very 60s and bizarre) been an architecture groupie, determined to ‘bed’ as many buildings as I could, starting with an under-funded trip to Barcelona to study Antoni Gaudi, resulting in a call home to my mother appealing for rescue from my penury.

It’s a very straight line, for me, to ponder the potential benefits of AI in the built environment and what ‘they’ would need to actually ‘Build better, Build Beautiful.’ So, I thought I would plunder my interests and cultural peccadilloes to wander through the notion of how robotics might inform the world of AI created architecture, even cities.

The question currently haunting all creative and intellectual effort – and of relevance here to the profession of urban architecture and placemaking – is whether AI has already slipped from a philosophical musing to become a true existential challenge.

As AI architects threaten to streamline or even replace the creative grunt work, one wonders if humanity is doomed to be slowly designed out of its own narrative.

By way of a tangential example, the newly developed AI Scientist from Sakana.AI provides a perfect case study. This is an AI machine that generates scientific papers at breakneck speed, capable of producing enough scientific knowledge to rival human contributions, all for a nominal cost of $15 per paper. This output of course lacks the critical element of creative suffering. The accidental innovations of penicillin, safety glass, Velcro, or even chocolate chip cookies were born of incidental chaos rather than structured perfection, and thus elude AI. And therein lies the opportunity.

Suffering as the missing element in AI

I am reminded of Masahiro Mori, a Japanese roboticist in the 70s, who coined the phrase ‘‘uncanny valley’—a phenomenon where robots or digital characters that closely resemble humans provoke feelings of unease or discomfort, as they appear almost but not quite human. In 1974, Mori published The Buddha in the Robot: a Robot Engineer’s Thoughts on Science and Religion in which he discussed the metaphysical implications of robotics. In the book, he wrote “I believe robots have the buddha-nature within them – that is, the potential for attaining buddhahood.” After his retirement from teaching in 1987, Mori changed his focus from engineering to something far less tangible and yet, paradoxically, far more human: he set out to explore the relationship between robotics and spirituality, especially through the lens of Zen Buddhism.

But this was no simple journey into the philosophical wilderness. Mori proposed that for robots to reach their true potential as extensions of humanity’s genius and folly alike, they must experience something deeper than mere functionality: they must suffer—not only to mimic humanity but to understand the gravitas of what is created. Inspired by Zen Buddhism, he envisioned “Zen-bots” (my phrase) – machines engineered to suffer, flipping the usual assumption that we want our technology to be endlessly optimistic. In a world that worships efficiency, Mori’s robots would comprehend constraints, be programmed to grasp loss and failure, and learn, in their way, to suffer. Mori’s approach to robotics through a Zen lens suggested that the quest for understanding machines mirrors deeper spiritual pursuits, such as overcoming the “illusion of control” between human and machine. His unique perspective reflects a harmonious relationship between technology and spirituality, emphasising that robotics could serve as a means to explore broader existential and metaphysical questions rather than merely functional or industrial goals.

His views, grounded in robotics research and writings from the late 1980s, echoed the Romantic vision of Mary Shelley, who saw technological creation as an act that brings inevitable torment to both creator and creature. Shelley’s Frankenstein warned us of the consequences of pushing science beyond ethical boundaries, and Mori’s Zen-bots tread this same uncanny valley, a mirror of humanity’s deepest contradictions.

Mori’s suggestion—and perhaps his warning—is that the most profound insights in human endeavour arise from struggle, mistakes, and persistence against obstacles. Without this element, a bot might produce flawless research, beautiful buildings, or seamless paintings, yet remain hollow, devoid of the chaos and grit that make its human counterparts come alive.

Mori’s Zen-bots hypothesis represents a truly existential turning point: machines that suffer as we do may, paradoxically, become better moral agents; robots that might serve humanity by understanding the “difficult whole” of the human experience.

Cyberpunk author William Gibson’s Neuromancer, who published in 1984, the first book in the Sprawl Trilogy, meanwhile imagines a world dominated by technology that, though sleek and powerful, lacks the gritty serendipity of human struggle. In that dystopian future, technology is disembodied and disconnected from humanity. Its AI—Wintermute —is bound by its programming and deeply limited in its self-awareness, almost like an intelligence caught within invisible bars. By contrast, Mori’s robots are infused with the limitations and burdens of sentience, embracing an echo of human mortality, an idea born not just from literature but from Mori’s formal studies in robotics and his exploration of Buddhist principles.

So, should we really make our machines suffer? Mori might argue that it’s the only way forward, that in doing so we avoid the sterile limits of mere utility. Just as Shelley’s monster longed for companionship, and Gibson’s Wintermute sought purpose, a suffering robot would gain an understanding of the ‘why’ of human existence. Until then, AI remains a mere apprentice, mimicking without understanding, always within arm’s reach of genuine creativity but never fully touching it. Suffering, Mori would suggest, could be the missing key that unlocks this potential—elevating robots from uncanny reflections to true participants in the human journey. Without the accidental discoveries that arise from ‘thinking outside the box’ — or God help me ‘blue sky thinking’ those messy, all-too-human moments — AI risks becoming a kind of Golem, mimicking but never truly understanding. A creation that serves no higher purpose or is built solely for pragmatic use may lack the essential qualities to function responsibly. In this way, the Golem reminds us of the ethical dimension necessary in all acts of creation: what we create is not just an object but a reflection of the values and intentions that drive it.

The future of AI as a co-creator in urban design

Urban resilience, at its core, is about cities that thrive despite disruption, ‘ or ’suffering’, where adaptability and empathy for the human condition lie at the heart of design. For AI to contribute meaningfully to architecture, it must understand the weight of resilience—the ability not only to construct but to transform spaces that respond to the intricate challenges and beauty of human life. Mori’s concept of a “Zen-bot” engineered to suffer hints at a powerful idea: if AI could comprehend struggle and adapt like humans, it might forge urban spaces that are not merely functional but deeply, emotionally resilient.

True resilience in architecture lies in human-centred design, where the needs, emotions, and even sufferings of communities are integral to the built environment. Imagine a “suffering AI” capable of empathising with human hardship, of intuitively understanding that the sterile uniformity often found in rapidly developed cities fails to meet our need for places of refuge, contemplation, and collective memory. Mori’s Zen-bots could offer a model here—AI systems that don’t just produce efficient layouts but sense the nuances of human experience. Such AI could help design spaces that respond to emotional needs, creating “pockets of resilience” in cities: quiet plazas, adaptable communal spaces, and sanctuaries from urban chaos.

Urban resilience demands structures and communities that are prepared for change both gradual and sudden. AI with an understanding of Mori’s Zen-bots’ ethos could aid architects and planners in developing designs that anticipate the unpredictable. These systems could simulate varied scenarios of change, such as climate-related challenges or social upheaval, and adapt designs accordingly. For example, buildings might be created with modular, flexible components that can be repurposed as community needs shift.

A resilient city is one that retains its character, quirks, and even its scars, qualities that human architects like Christopher Alexander celebrate in his book Pattern Architecture. An AI informed by Zen-bot suffering could reject the allure of sterile, homogenous designs and instead honour the beauty of imperfection.

Already, AI is influencing architecture, with generative design algorithms creating innovative solutions like Zaha Hadid Architects’ fluid, parametric organic structures. However, these designs risk appearing distant from the human experience unless they can resonate on a deeper level. Imagine if Mori’s Zen-bots approached architecture through the lens of resilience: they might create adaptable, living spaces that change with seasons or societal needs.

At its most profound, urban resilience is about crafting spaces that are emotionally as well as structurally robust, cities that echo the trials, losses, and triumphs of their inhabitants. Resilient cities are not just efficient but deeply human—they offer spaces that evoke joy, contemplation, and connection. With “suffering” AI, urban design could become a collaborative narrative, where robots help tell humanity’s stories, honouring both our fragility and resilience.

As we contemplate Mori’s Zen-bots and their potential for “creative suffering,” the stakes in architecture grow clear: do we allow AI to generate spaces stripped of the human spirit, or do we challenge it to reach beyond the uncanny valley into a realm of empathy? Cities aren’t just Corbusian efficient machines; they’re reflections of humanity’s beauty, failures, and resilience. The integration of AI in urban design could either echo Mori’s vision of a machine that struggles and grows, or it could lead to soulless, homogenous landscapes where humanity’s voice is a faint whisper.

Cities are, fundamentally, more than functional structures; they are vessels of collective memory, culture, and identity.

This ‘human layer’ encompasses the beautiful, the flawed, and the resilient — all essential facets that make cities what they are. Streets are repositories of community histories, architecture speaks to aesthetic and cultural ideals, and public spaces are arenas where social interactions unfold. These aspects do not simply meet a functional need but cultivate a sense of place, which is often absent in purely functionalist design.

Cities also bear the visible scars of human failures: neighbourhoods affected by disinvestment, pollution, crime, and socioeconomic divides. Such challenges might be seen as evidence of human error, but they’re also powerful symbols of resilience. Communities adapt, resist, and rebuild; the presence of derelict sites transformed into lively, community-centred spaces exemplifies this resilience. Moreover, historic preservation and adaptive reuse initiatives reveal a growing movement to revitalise rather than replace, creating beauty from remnants of the past.

True urban architecture does not just organise space; it curates experiences, trials, and triumphs that connect us. If we train AI merely to imitate without comprehending suffering, we reduce the city to an inert playground, devoid of the grit and imperfection that gives it life. Yet, by embedding our struggles, successes, and philosophical journeys within AI, we might just find ourselves co-creating urban spaces where robots become partners, not merely tools—places where both humans and machines experience the shared rhythms and aspirations of urban life, evolving together in ways that enrich the very fabric of our cities.